heterogeneity of exposure and attribution of mesothelioma: … · 2015-01-28 · heterogeneity of...
TRANSCRIPT
This content has been downloaded from IOPscience. Please scroll down to see the full text.
Download details:
IP Address: 132.206.227.34
This content was downloaded on 28/01/2015 at 21:33
Please note that terms and conditions apply.
Heterogeneity of exposure and attribution of mesothelioma: Trends and strategies in two
American counties
View the table of contents for this issue, or go to the journal homepage for more
2009 J. Phys.: Conf. Ser. 151 012008
(http://iopscience.iop.org/1742-6596/151/1/012008)
Home Search Collections Journals About Contact us My IOPscience
Heterogeneity of exposure and attribution of mesothelioma:
Trends and strategies in two American counties.
B W Case1,2
and J L Abraham3
1. Department of Pathology and School of Environment, McGill University, Montréal,
Québec, Canada. Email: [email protected]
2. Professeur invité, Epidemiology & Biostatistics Unit, INRS-Institut Armand-Frappier,
Université du Québec, Laval, Québec, Canada.
3. Department of Pathology, SUNY Upstate Medical University, Syracuse, NY 13210,
USA.
Abstract. As mesothelioma risk has begun to decline in the United States, two trends are gaining
relative importance. ―Legacy‖ exposures causing this disease are most important in locales having
past asbestos industry, shipyards, and/or local distribution of asbestos amphibole-containing
material as a result. ―Future‖ exposures are of particular concern in relation to so-called ―naturally
occurring asbestos‖ (NOA) areas which include unequivocally asbestiform amphibole. In this
paper, Jefferson Parish, Louisiana is used as an example of the first trend, and El Dorado County,
California as an example of the second. Available tumor registry, epidemiology, historical and
mineralogical data, and lung-retained fibre content are used as indicators of disease and exposure.
Jefferson Parish, LA was chosen as the prototype of ―legacy‖ exposures on the basis of historical
evidence of asbestos plants with known mesotheliomas in the workforce, known shipyards in the
same area, EPA records of distribution of crocidolite-containing scrap to and remediation of over
1400 properties, NIOSH published data on mesothelioma by county, and exposure data including
lung-retained fibre analyses in victims, where available. El Dorado, CA was chosen as the
prototype of NOA amphibole exposures on the basis of tumor registry data, activity-based EPA
sampling data in one area, and lung-retained fibre analyses in area pets, and future risk assessment
based on tremolite-specific modelling in Libby, Montana and elsewhere. As expected, the legacy
exposure area was high in mesothelioma incidence and mortality. Lung-retained fibre content
confirms crocidolite exposures in exposed plant-workers and those exposed to crocidolite-
containing scrap, and amosite in shipyard workers. In contrast, to date, cancer registry data in the
NOA-amphibole (―future‖) county does not show a clear increase in incidence or mortality, but
grouped county data from the area show a shift in higher incidence rates to the NOA areas and
away from California ―legacy‖ (e.g., shipyard) areas from 1988-2005. EPA active sampling has
confirmed excess tremolite/ actinolite fibre(s) in air, although there is debate about its nature and
the appropriateness of the area sampled. Lung-retained fibre in local pets shows unequivocally
elevated asbestiform tremolite/ actinolite in areas thought to be most affected, but numbers are
small. Future risk is expected to rise due to a vastly increased population base coupled with
Inhaled Particles X, (23–25 September 2008, Manchester) IOP PublishingJournal of Physics: Conference Series 151 (2009) 012008 doi:10.1088/1742-6596/151/1/012008
c© 2009 IOP Publishing Ltd
1
exposures potentially created by related construction activities. Although legacy exposures are
producing smaller numbers of cases with time, they continue to occur at high rates, and new
sources of legacy exposure are being discovered in highly localized ―hotspots‖. Differential
exposure sources remain a problem in attribution, but continued remediation seems the best
strategy for prevention. In the ―future‖ risk county and surrounding areas, incidence trends are less
clear, but again highly localized exposures as opposed to broad areas seem important. Activity-
based air sampling; targeted soil samples, and lung-retained fibre analyses may be useful in
defining areas of highest future risk and potential prevention.
1. Introduction
Patterns of geographical distribution of mesothelioma cases in the United States were last studied in depth
over 20 years ago [1,2]. Authors noted that, with some exceptions, highest concentrations of cases were
likely to be noted in counties having either asbestos-using industrial plants, shipyards, or both. Most
studies of mesothelioma incidence in the United States have relied on the Surveillance, Epidemiology, and
End Results Program (SEER) data, a subset which is incomplete but which also supplies cases for nine
jurisdictions [3]. According to this source of information, mesothelioma incidence among U.S. men
peaked before the turn of the century and has begun a slow decline [3,4], possibly due to decline of use of
commercial amphiboles since the late 1960’s [3].
While the decline is welcome, it brings new challenges for mesothelioma ascertainment. These can be
divided into three parts:
a deficit in the SEER data itself
a persistence of the ―legacy‖ exposures both in their traditionally known forms (as in
continuing risk from past direct and indirect exposures from transformation plants and
shipyards) and newly discovered offshoots (as in the distribution of asbestos-containing (AC)
scrap from such sites), and
exposure from mineral deposits not commercially exploited, either directly via disturbance of
the material in situ or indirectly via transport by wind or water; this is sometimes called
naturally occurring asbestos (NOA), although the distinction is in lack of present commercial
use, not geological occurrence. Past sources of commercial exploitation can become current
sources of NOA exposure, or they can occur through weathering or development not
specifically aimed at extraction of the minerals, such as construction activities or real-estate
development siting.
1.1. Deficits in the SEER data itself.
SEER data is by definition limited to those jurisdictions from which it is derived. Both overestimates and
underestimates of mesothelioma incidence nationwide are theoretically possible, depending on the
representativeness of the data. It has been suggested that, for example, studies of attributable risks of
varying causes of mesothelioma based on the SEER data may have overestimated shipyard, ship-building
and ship repair, since the data-set includes the Los Angeles County area (as well as cases and controls
from the Veterans Administration). Los Angeles County had a disproportionate share of such activities
and 46% of incident cases were attributable to exposures from those sources there, not ―directly
representative of the whole population of the United States‖ [5]. The accuracy of diagnosis data within
the SEER data has also been questioned, but this incidence data matches national mortality data quite well
for 1998-1999 if compared with ICD-10 mortality data for mesothelioma for the same areas [6].
Inhaled Particles X, (23–25 September 2008, Manchester) IOP PublishingJournal of Physics: Conference Series 151 (2009) 012008 doi:10.1088/1742-6596/151/1/012008
2
1.2. Persistence of legacy exposures.
Commercial amphibole exposures occurred long after the late 1960’s in the United States. For example,
crocidolite consumption in the manufacture of products constituted nine per cent of U.S. total asbestos
consumption in 1981, 95% of which was in the production of asbestos cement pipe [7]. Amosite
consumption in insulation production continued to a lesser degree after 1972, when the U.S. Navy
formally discontinued asbestos use (amosite had constituted up to 85% by weight, by specification, of
naval ship insulation during World War II [8]). However, exposure to amosite in installed insulation and
other products in secondary trades continued (and continues), as amosite by one account was identifiable
in the lungs of 82% of all mesothelioma victims analyzed over the last 25 years [9].
An emerging source of exposures to commercial fibre is domestic use of industrial scrap, which unless
painstakingly remediated may remain in a neighborhood virtually forever. This has occurred for example
in Italy near a crocidolite-using asbestos cement plant, where close examination of exposures of
mesothelioma cases identified ―the use of AC residuals such as mixing them into the soil to create hard
pavement and improve water absorption or applying layers of finely ground AC—rich in asbestos fibres—
in the lofts for thermal insulation‖ [10]. Distribution of crocidolite-containing scrap throughout an entire
county by an American cement pipe plant is an example offered in this paper, for Jefferson Parish,
Louisiana; a county which included other asbestos product plants and shipyards as well. Exposures from
all three ―legacy‖ sources have contributed to the burden of mesothelioma, making it one of the highest-
incidence counties in the nation (see ―Results‖).
1.3. Exposure from mineral deposits not commercially exploited (NOA).
Of the 50 American states, 35 have been identified by the United States Geological Survey (USGS) as
containing asbestos occurrences, most of which have not been exploited. These are largely in two areas.
The first begins in Eastern Canada and runs down the eastern coastal and inland states; USGS counts 60
former asbestos mines and 331 ―Natural asbestos occurrences‖ in this area, including not only chrysotile
but former anthophyllite asbestos mines in Georgia, North Carolina, Virginia, Maryland, Connecticut,
and Massachusetts [11]. The second most important belt of occurrences runs from Alaska through British
Columbia, Washington State, Oregon and California; as yet, it has not been as well characterized although
California was the site of the largest chrysotile mining operations in the western USA and chrysotile was
also mined in northern British Columbia. Other occurrences have been identified in the central and
―Rocky Mountain‖ states [12,13]. Most recently, USGS has published tables and maps of NOA in the
southwest including Arizona, Utah and Nevada, including information on 113 occurrences, most in
Arizona where there are 46 former chrysotile mines, the last of which closed in 1982 [14]. Tremolite,
apart from being possibly associated with occurrences of other minerals, was also specifically mined in
Utah; tremolite and actinolite are commonly encountered but not well characterized as to whether they are
―asbestiform‖ or ―non-asbestiform but fibrous‖, a distinction which may lack biological significance [15].
Except for the unusual Zonolite Mountain deposit of mixed asbestiform amphibole at Libby,
Montana, which had a very high proportional mortality for mesothelioma among the workforce [16-18]
and residents [18] in and near the vermiculite mine located there, there have been no real investigations of
NOA mesothelioma causation – or even of the workforces of the over 100 previous asbestos mines. A
single exception looked at residential proximity to California asbestos deposits as the independent variable
for mesothelioma incidence 1988-97 in the California Cancer Registry, and found a significant
relationship [19]. Like all ecologic studies this one was potentially hampered by inadequate occupational
and other individual exposure data, including residential proximity itself other than that at time of
diagnosis, but the positive results do suggest caution in residential siting. El Dorado County, on the
western slope of the Sierras and known to contain deposits of asbestiform tremolite/ actinolite, has been an
area of particular concern in this regard and is the county used as the NOA example in this paper.
Inhaled Particles X, (23–25 September 2008, Manchester) IOP PublishingJournal of Physics: Conference Series 151 (2009) 012008 doi:10.1088/1742-6596/151/1/012008
3
2. Methods
2.1. Selection of counties and sources of information.
The choices of Jefferson Parish, Louisiana as representative of a ―legacy exposure‖ county and of El
Dorado, California, as representative of an ―NOA exposure‖ county were made a priori on the basis of
familiarity of the authors with these areas. As a first step in seeking out information, state cancer
registries in the two areas were examined (although as noted below this proved unsatisfactory in
Louisiana). Historical exposure data, including published papers, geological maps, and documentation of
exposure in the two areas by regulatory agencies was sought out in general terms, including identification
by bibliographic search (e.g. Pubmed). Individual mesothelioma cases were culled from the files of one of
us (BWC) for Jefferson Parish; neither of us had a file on a case in El Dorado county, which we regard as
an area primarily of future risk rather than past exposure, although certainly exposures may have occurred
in the past. Information on potential exposure(s) was obtained from deposition testimony, social security
records, medical records, and records of employment where available. Where possible, lung tissue was
obtained and analysed as outlined in the following section.
2.2. Lung-retained fibre analysis.
Mesothelioma cases from Jefferson Parish for which lung-retained fibre analysis were performed by one
of us (BWC) were analysed according to the methods published for the McGill University laboratory [20-
25]. Briefly, lung samples were obtained from paraffin blocks after identification from corresponding
stained slides, deparaffinized with filtered xylene, and prepared for analytical electron microscopy
following digestion with filtered household bleach followed by oxidation in a low-temperature asher with
grid preparation using a carbon replica technique. Fibre identification and counting were performed at a
magnification of 13,500X in a Phillips 420™ electron microscope equipped with an energy dispersive
spectrometer permitting the capture of X-Ray energy dispersive spectra (EDS) of individual particles.
Fibres were defined and counted as having length greater than 5 um and aspect ratio greater than 3:1.
EDS spectra were observed and evaluated online; reference standards for comparison were spectra
obtained from similarly prepared grids holding UICC samples of amosite, crocidolite, and chrysotile and a
Research Triangle Park sample of tremolite, among others. Fibres were counted in two samples from
different parts of lung to a theoretical maximum of sixty fields per sample (detection limit 34.37 fibres /
mg dry lung).
For El Dorado county, no mesothelioma cases were available: for exposure assessment cat and dog
lungs obtained from local veterinarians were obtained and analysed using both the McGill methods as
above and the SUNY lab methods. The latter were similar with respect to sample preparation and fibre
identification, but scanning electron microscopy was used at a magnifications of 4000X and 8000X rather
than transmission electron microscopy as in previous work from that laboratory [26-28].
3. Results
3.1 Legacy exposures in Jefferson Parish, Louisiana.
The Louisiana Tumor Registry proved not to be a useful source for county-by-county comparisons of
mesothelioma rates. This is because that registry groups counties together into regions for some purposes,
and therefore Jefferson Parish – the chosen county for this paper – is grouped with Orleans Parish (which
includes the city of New Orleans proper) as well as St.-Bernard Parish. The latest figures for the three
counties combined nonetheless show the region to have the highest age-adjusted (to the 2000 U.S.
Inhaled Particles X, (23–25 September 2008, Manchester) IOP PublishingJournal of Physics: Conference Series 151 (2009) 012008 doi:10.1088/1742-6596/151/1/012008
4
population) annual incidence rate in the state at 2.1 per hundred thousand for the five years from 2000-
2004, based on 104 recorded cases (69 in white males; 19 in white females), as compared to 1.3 for the
state as a whole [29]. The county was not listed as one of the high mesothelioma rate counties in 1987
[1,2], and is not included in SEER data (nor is any county in Louisiana), but the New Orleans ―state
economic area‖, which includes Jefferson Parish, had a ―significantly high‖ SMR for ―pleural cancer‖
(ICD8 code 163.0) among white females for U.S. death certificate data 1968-78, although not for males
[2]. Overall, 14 of 174 such economic areas in the US had ―significantly high‖ female pleural cancer
mortality, and of these five, including the New Orleans area, had both shipyard(s) and asbestos
manufacturing plant(s). At least two of the latter were present in Jefferson Parish; workforces for these
were studied by Hughes and Weill [30]. One of the two was the Johns Manville plant, which included a
pipe-manufacturing facility which used crocidolite ―steadily‖ in manufacture of ―transite‖ cement pipe.
Hughes et al. identified eight mesothelioma cases in workers from this plant, and found further in an
internal case-control analysis that mesothelioma was related to proportion of time spent in the pipe area, as
well as duration of employment. Although the plant existed from 1928, the pipe plant opened in 1947.
According to US EPA and Louisiana Department of Environmental Quality (LDEQ) documents obtained
through freedom of information requests, through the 1950’s and 1960’s, in addition to atmospheric
pollution from the plants, plant personnel produced an asbestos-containing aggregate which was
distributed throughout the communities of the surrounding area (the ―Westbank‖ of the Mississippi River).
This contained both chrysotile and crocidolite asbestos, in a proportion varying from 1:3 respectively to
1:1, and determined by testing to be 35% to 45% total asbestos by weight. It was created in a hammer
mill and mixed with a filler, usually composed of gypsum, dolomite, or calcite. The asbestos aggregate
and filler formed ―a concrete-like material when mixed with water and therefore was considered by many
local residents to be a concrete substitute for construction purposes‖ [31]. It was distributed to local
residents free of charge, often by pickup or dump trucks. It was first identified as a potential hazard in
1990; then again in 1992 and 1994, but by 1996 it had become friable and was considered an ―imminent
threat‖ to public health, with mesothelioma mentioned specifically as a possible outcome from exposures
[31]. The material was ―located in residential yards and driveways, school playgrounds, around day care
centers, and in other areas easily accessed by the public‖ and ―Children were seen playing on driveways
composed of friable ACM with toys and basketballs…vehicles were observed creating dust clouds when
passing over areas that contained the ACM. All of these routine activities (were) expected to increase
ACM friability and dramatically increase human exposures‖ [31]. In 1996 remediation activities began,
and currently over 1400 properties have been remediated; however, a walk-through of the residential areas
near the plant in 2008 by one of us (BWC) clearly identified areas without mitigation, including some in
the drainage paths from the plant and from dump sites. In 1999, according to NIOSH data, Jefferson
Parish had the eleventh-highest age-adjusted mortality rate at 39.9 per million population for U.S.
residents age 15 and over, based on six deaths in women and eight in men (the corresponding overall U.S.
rate ranged between 11.32 and 11.64 in the four years from 1999-2002) [32]. Recently NIOSH published
an up-to-date supplement of this data for the five years from 2000 to 2004, finding 75 deaths from
mesothelioma in that time period (age-adjusted rate 42.1 deaths per million, of which 34.7% were female)
in Jefferson Parish [33]. This is by far the largest current number of mesothelioma victims in high-rate
counties in the USA, exceeding by fifteen-fold for example the five deaths seen in the same time period in
Lincoln County, Montana, where the Libby ―epidemic‖ is situated [33].
3.1.1. Total cases identified. Twenty-two Jefferson Parish medicolegal mesothelioma cases (ten female
and twelve male) were identified by one of us (BWC) from our files. At least six of the 22 were related;
two sets of siblings and one parent and child. All were resident at some time in their lives (most
frequently for most of their lives) in Jefferson Parish. All had pathologically verified mesothelioma,
Inhaled Particles X, (23–25 September 2008, Manchester) IOP PublishingJournal of Physics: Conference Series 151 (2009) 012008 doi:10.1088/1742-6596/151/1/012008
5
confirmed with immunohistochemistry. All but one were pleural. The peritoneal case was a 39 year old
female with no definite history of asbestos exposure. All other cases (12 male and 9 female) were pleural
and had a definite history of asbestos exposure including one or more of direct occupational exposure in
asbestos manufacturing plant(s) or shipyard(s) in the area; indirect exposures (household or domestic
exposure, usually via the laundering of clothes from such workplaces by family members); or
environmental exposures through living near asbestos manufacturing plants or at home addresses where
asbestos-containing materials had been directly deposited. Some had also attended schools where the
ACM scrap was placed in the yards during childhood. Many had suggested possible exposures from
multiple sources including two or more of the above, making attribution difficult in some cases.
3.1.2. Abbreviated exposure histories for pleural mesothelioma cases for which lung tissue analysis
could not be performed. For 12 of the 21 pleural cases (six men and six women), no lung tissue was
available for assessment of retained fibres to aid in exposure assessment. Of these, four were men who
worked in shipyards; two for at least twenty years, one in his ―college years‖, and one before becoming a
fireman. The latter recalled exposure to amosite-containing insulation and many other asbestos-containing
products and also had a crocidolite scrap driveway as a child. A fifth man had crocidolite-containing
scrap material deposited at his residence in childhood; a sixth had such material in the neighborhood but
could not recall his childhood address. Both of the latter also claimed chrysotile exposure in automotive
brake work. Of the six women with no lung fibre analysis five had crocidolite-containing scrap deposited
on their property as children, and one of the five lived approximately three hundred feet from the pipe
plant for twenty years. Four laundered clothes of fathers, husbands and/or sons who worked at one time
either in the asbestos manufacturing plants in the area, in shipyards, or in both.
3.1.3. Cases identified with previous fibre analysis or with lung tissue available. Lung tissue had been
analysed previously in other laboratories in two cases. One was a 66 year old woman who had a
crocidolite-containing scrap driveway as a child and one crocidolite-cored asbestos body and two
tremolite fibres were identified from lung tissue; the other was a shipyard electrician who had an
unspecified number of amosite fibres identified in his lung tissue with relatively low (4000 amosite fibres
per gram wet lung) concentrations. Close examination of the methods in the latter case however suggest
this result may not be reliable, as the analyst had no past experience with the technique, no description of
methods and no published or unpublished control values. Samples were available for our own analysis
and provided in seven cases. Results for these cases are shown in Table 1.
Inhaled Particles X, (23–25 September 2008, Manchester) IOP PublishingJournal of Physics: Conference Series 151 (2009) 012008 doi:10.1088/1742-6596/151/1/012008
6
Table 1. Jefferson Parish mesothelioma cases with lung-retained fibre analysis
a,b.
Age, Sex
Latency
Primary
exposure
Other
exposures
Crocidolite Amosite Otherc
49 M;
49 yrs
Residentiald
Barges (no
insulation)
344, 344
n.d.e, n.d. n.d., 34 (Ch)
2640 (AB)
64 M;
42 yrs
Shipyard
electrician
Residential n.d. 3093, 2690 Fibres n.d. by TEM;
15,360 (AB)
52 F;
52 yrs
Residential none 172, 206 n.d. Fibres n.d. by TEM;
1480 (AB)
57 M;
52 yrs
Residential Sales, home
improvement
n.d. n.d. Fibres n.d. by TEM;
AB n.d. by PCOM
60 F;
54 yrs
Residential Domesticf
309, 109 n.d. n.d., 69 (Ch)
8000 (AB)
75 M;
58 yrs
Occupational-
Pipe plant
Occupational-
Chem. Plant
722, 1134 69, 137 34, n.d. (Ch); n.d., 69 (T)
24,000 (AB)
57 M;
52 yrs
Residential Shipyard,
< 1 yr.
1134, 825 n.d. No other fibres by TEM;
2933 (AB) a Fibres > 5 micrometers per mg dry lung; aspect ratio > 3:1; detection limit 34 f/ mg (i.e. 1 fibre = 34 f/mg).
b Where two values are given, two separate analyses were performed from two parts of lung.
c (Ch) = chrysotile (background = 60-80 [15, 20-22]); (T) = tremolite (background = 40-60[15,20-22]);
(AB) = asbestos bodies by phase-contrast optical microscopy per gram dry lung (background < 1000). d Residential = crocidolite-containing scrap at residence in childhood.
e n.d. = none detected
f Alleged household exposure via laundering of clothes of relative who worked in shipyard < 2 yrs
3.2. Assessment of NOA exposures in El Dorado County, California
3.2.1. Previous work by others. The possibility of tremolite/ actinolite asbestos exposures in the then
thinly-populated county of El Dorado, east of California’s capital Sacramento, first came on the national
scene with a series of newspaper articles by Chris Bowman of the Sacramento Bee. Mr. Bowman was
following up on the dilemma of a local citizen who had felt forced to leave his home in the county because
of the newly-discovered risk. It has long been known that such fibre was present; it is clearly indicated for
example on the 1970 state geological maps used by Pan et al. [19] in their assessment of mesothelioma
risk and proximity to naturally occurring asbestos deposits. In an intraperitoneal animal experiment
designed to highlight differences in risk between ―asbestiform‖ and ―nonasbestiform‖ tremolite/ actinolite
(the latter has not been regulated as asbestos since 1992 and there is debate about health effects of the very
wide range of fibres and particles covered by the definition[15]), the asbestiform fibre found to produce
mesothelioma most rapidly and consistently was one from a source near this area of California [34].
Citizen concern in the area sparked regulatory agency attention, with an activity-based sampling regimen
applied by US EPA in a study performed in 2005 [35]. During the study, EPA technical personnel
simulated recreational activities at schools and a park, including cycling, team sports and running.
Personal air samplers were worn during the activities and dust was collected from breathing zones to
Inhaled Particles X, (23–25 September 2008, Manchester) IOP PublishingJournal of Physics: Conference Series 151 (2009) 012008 doi:10.1088/1742-6596/151/1/012008
7
represent potential exposures to both children and adults. Several protocols were used in electron
microscopic fibre counts, including modified ISO 10312; ―Phase Contrast Microscopy Equivalent‖
(PCME) counts of fibers longer than 5 micrometers, with aspect ratios greater than 3:1 (to conform to
measures used in exposure assessment in historical epidemiology studies using phase contrast optical
microscopy). Results were summarized by EPA: ―U.S. EPA found asbestos fibers in almost all El
Dorado Hills air samples, including those collected nearby, but outside the area of activity. In general,
personal asbestos exposures from simulated sports and play activities were significantly elevated over
levels observed in the nearby asbestos air samples taken outside the area of activity. The dominant
asbestos fiber type detected was amphibole, which is considered to be more toxic than chrysotile‖ [35].
There was some debate between EPA, an industry lobbying group, and some elements in the El Dorado
community governance about the appropriateness of counting methods, and the degree to which identified
amphiboles were asbestiform (as opposed to ―cleavage fragments‖); leading to the US Geological Survey
conducting their own investigation in the area [36]. They concluded that both types of ―fiber‖ were
present at different sites, and that ―Based on aspect ratio data of the less than 3 um diameter particles, the
morphologies of the amphiboles in the El Dorado Hills area are intermediate between what might
generally be considered a population of commercial-grade asbestos particles and a population of cleavage
fragments, produced by milling of massive amphibole‖ [35]. They stressed further that measurement, not
health effects evaluation of any population of fibres, is the province of mineralogists and allied scientists:
―It is the obligation of the analytical and mineralogical communities to provide accurate, unbiased, and
scientifically sound information to the health and regulatory communities so that appropriate and
informed, health-related policy and regulatory decisions can be made‖ [36].
This series of events left many in the community confused, divided and angry. Some saw a playing
down of risk based on obscure mineralogical criteria with no proven health implications. Conversely,
others saw unfair singling out of their community (California being well-known to contain much
serpentine rock – indeed, it is the ―state rock‖ – often home to asbestos deposits which are largely
chrysotile), even though the instigation and emphasis of the investigations had always been on the basis of
the presence of amphiboles, and any arguments between scientists had had no relationship to chrysotile
asbestos). Although further studies had been planned, including government-sponsored ―expert‖
conferences and risk assessments, none were performed, leaving the situation in limbo.
3.2.2. California Cancer Registry Data. We looked at figures in the California Cancer Registry using
their online tool, which allows county-by-county comparisons for any time period demarcated by years
between 1988 and 2005. Given the small population at risk, long latency for mesothelioma and time
window we expected this might not be sufficient for detecting trends. Mesothelioma (ICD-10) was not
coded but pleural cancer was. In fact, there were 33 cases registered for ―Pleural Cancer‖ from 1988-2004
for an age-adjusted rate per 100,000 of 1.36; seventh highest in the state. Also of interest, two of the six
higher-incidence counties (sometimes grouped) were immediately adjacent and also on the Western Slope
of the Sierra, where tremolite/ actinolite deposits are scattered: Mariposa – Toulomne (30 cases; rate 1.93)
and Alpine - Amador – Calaveras (23 cases; rate 1.86). For comparison purposes, Los Angeles County,
included as noted above in SEER data and having in the past 46% of cases attributable to exposure from
shipyards [2] had a rate of only 0.85. However, this was based on 1044 cases 1988-2004, with a
―population at risk‖ two orders of magnitude higher. Nevertheless, 95% CI on rates did not overlap,
showing a significantly increased mesothelioma rate for El Dorado County vs. Los Angeles County (and
for the other Western Slope counties) over this time period (difficulties with interpretation of this type of
comparison are evident and are discussed later).
Inhaled Particles X, (23–25 September 2008, Manchester) IOP PublishingJournal of Physics: Conference Series 151 (2009) 012008 doi:10.1088/1742-6596/151/1/012008
8
3.2.3. Lung-retained fibre analysis. It has not been possible to obtain lung samples from mesothelioma
victims – or from hospitals, coroners’ offices, or any other source – in El Dorado County, despite four
years of efforts in that direction. In the interim, we elected to begin to collect as possible pets (cats and,
preferably, dogs) who died of natural causes in county. These were provided by concerned citizens and
lungs were obtained in toto at post-mortem by local veterinarians and supplied to our laboratories for
analysis. To date, we have received lung samples from two cats and four dogs; five have been analysed in
both laboratories, as reported here; analysis is not complete in the McGill laboratory for the final dog.
Results for the from both McGill and SUNY laboratories to date for tremolite/ actinolite concentrations at
two fibre lengths are shown in Table 2.
Table 2: Lung fibre concentrations in El Dorado County pets from two laboratoriesa:
Tremolite/ actinolite fibres/ gram dry lung, aspect ratio > 3:1
ANIMAL Exposure SUNY Lab
> 5 um length
McGill Lab
> 5 um length
SUNY Lab
> 10 um length
McGill Lab >
10 um length
Cat 1 none 0 0 0 0
Cat 2 9 years,
indoors
86,000 157,000 86,000 157,000
Dog 1 > 2 years 1,250,000 566,500 192,000 412,000
Dog 2 > 8 years 9,162,000 928,000 2,030,600 928,000
Dog 3 unknown 186,000 204,000 72,000 47,500
Dog 4 unknown 961,000 - 448,000 - aMcGill Laboratory: each value is the mean of two samples; analysis by Transmission Electron
Microscopy (TEM) ; SUNY (State University of New York) Laboratory: each value is the mean of
two to three samples; analysis by Scanning Electron Microscopy (SEM). Both laboratories used
similar sample preparation and fibre identification for tremolite/ actinolite (by EDS) methodology.
Of the six animals shown, the first (Cat 1) did not live in an area known to contain tremolite/ actinolite.
The second cat did, but (like most cats) spent most of its life indoors. Dogs 1 and 2 both lived in
―exposed‖ areas but as shown there was greater lung concentration of tremolite/ actinolite shown in both
laboratories for the dog with longer residence time. For the third and fourth dogs we have not yet
obtained ―residential histories‖, and analysis is incomplete. Other fibers were also found but in small
numbers and concentrations except for Dog 3 in the McGill Laboratory; this dog had 204,000 fibres (mean
of two counts from two portions of lung based on eight fibres seen) tremolite/ actinolite with mean length
10.4 micrometers but chrysotile concentration was, unlike in the other cases, very high at mean over 2.5
million fibres per gram dry lung and very long with mean fibre length over 13 micrometres. However,
this was not replicated in the SUNY laboratory in which no chrysotile was seen at a detection limit
varying between 7000 and 20,000 fibres per mg dry lung.
4. Discussion
Results in these two counties show two very different patterns relating to exposure and mesothelioma risk.
Each presents scientific and medical dilemmas; each can be aided by obtaining accurate statistics on
mesothelioma incidence and mortality, and ideally, lung-retained fibre analysis (where possible). In this
section we discuss the central dilemmas for legacy exposures and for future risk from NOA exposures and
suggest ways in which addressing them may be improved, as well as current deficiencies both in the
available data from these counties and in our own analysis. Finally, addressing the different problems for
these two situations has some common – and common-sense – solutions.
Inhaled Particles X, (23–25 September 2008, Manchester) IOP PublishingJournal of Physics: Conference Series 151 (2009) 012008 doi:10.1088/1742-6596/151/1/012008
9
4.1. Legacy exposures
Although asbestos use (consumption) in the USA peaked in 1973 [37] and mesothelioma incidence is said
based on SEER data to have peaked in the 1990’s [3,4] U.S. mesothelioma mortality remains
approximately flat at 2500 cases per year for the most recently available (1999-2002) National Center for
Health Statistics multiple cause-of-death data (ICD-10 data). As may be seen in this report, latency
remains very long, although there is some indication that the linear increase with time from first exposure
after approximately 20 years flattens out after 50 years from first exposure [38]. It is therefore not
surprising that most cases continue to be due to legacy exposures; indeed on the basis of consumption
figures alone incidence would be expected to keep rising for another decade. New trends have appeared
in addition to the known exposures, which in the USA were maximal in counties that had asbestos
manufacturing plants and shipyards. Some legacy exposures are due to secondary exposure from the
latter, both through past neighborhood and domestic exposures and past and continuing exposure from
material distributed from manufacturing plants, such as that described here for Jefferson County,
Louisiana. The exposure profile overall in such a neighborhood can therefore historically be quite
complex, leading to concomitant complexity in attribution. The latter is relatively easy to estimate in a
population by, for example, type of exposure or occupation. Attribution of the individual case, however,
sometimes called ―specific causation‖, remains a problem. One approach has been to apply what we
know about population risks – namely, that they are related to differential fibre potency [39] and to the
third to the fourth power of time from first exposure [40]. One model has been constructed to apply dose
reconstruction for these two powerful factors in general causation to specific causation [41]. While this
has some appeal, it is controversial. For one of us (BWC) it lacks biological rationale for the individual
case, since it does not seem possible given the accepted tenets of carcinogenesis that in the individual case
a large number of exposures could simultaneously, or even in sequence, have been causal. On the other
hand, experts disagree on this matter, and indeed that includes the two authors of this paper. One of us
(JLA) feels it is, conversely, quite biologically plausible that ALL exposures contribute to the resultant
mesothelioma, and that it is not currently possible to exclude any one documented exposure completely, if
had that one lesser exposure been the only one, there would be no argument that it was sufficient and the
only cause.
Whichever view is taken, this leaves those concerned with compensation for the individual case with a
dilemma – when more than one exposure has occurred, how does one decide biologically which
exposure(s) is (or are) causal? In many jurisdictions such a decision must be made, and some US Appeal
Courts do not allow ―expert‖ opinions with wordings such as ―each and every exposure contributed to…‖
or ―each and every breath of asbestos-containing air was a cause of…‖ the disease.
While it appears sensible, where possible, to assign attribution to the most important exposure (in
terms of dose or some surrogate for dose, with the above-noted corrections for fibre type and for time
from first exposure), it is not always possible. Evidently, if science cannot determine which exposure is
the one which was causal in the individual case, and multiple possibilities exist, there needs to be a
balance between fairness to the individual with disease and fairness to those, if any, deemed ―responsible‖
for the exposures that constitute those possibilities. Where available, lung-retained fibre analysis can
provide one component of the solution to this difficulty. This is evident in the Jefferson Parish cases
having unequivocal lung-retained fibre results in which it is clear either that a professional (for example,
amosite from shipyard insulation or crocidolite from cement pipe manufacture) or residential (for
example, from crocidolite-containing scrap on a person’s property to which that person was exposed in
childhood) exposure was most significant. Arguments may be raised that rapid differential clearance may
cause an exposure to (for example) chrysotile to ―disappear‖; conversely however it is just this rapid
clearance that has often been raised as the explanation for the wide differences in risk between fibre types
Inhaled Particles X, (23–25 September 2008, Manchester) IOP PublishingJournal of Physics: Conference Series 151 (2009) 012008 doi:10.1088/1742-6596/151/1/012008
10
[3, 39, 41]. Of course, lung tissue is in any case not always available from medical procedures such as
diagnostic lung biopsy, therapeutic pleuropneumonectomy, or autopsy requested by relatives.
Better cancer registries – or preferably, a dedicated mesothelioma registry of the type available in parts
of France [42] and in the past (1980-85) in Australia [43] – could contribute as well, although their main
contribution would be to a better ascertainment, diagnosis, and treatment of the disease. Unfortunately the
Australian example with the demise of the quality of the registry with defunding after 1985 [43] indicates
that obtaining adequate funding for this endeavour will not be a simple matter. Pending legislation in the
US Congress has suggested funding a Registry there, but the details remain vague.
No matter what solution is suggested, there will inevitably be controversy associated with it, as fixing
blame on a maximum number of potential parties increases potential rewards in tort litigation both for
patients and lawyers, while also spreading financial risk more thinly among defendants.
4.2. Future risk from NOA exposures.
As with legacy exposures, exposures such as NOA, which create future risk, may be aided in improved
assessment via cancer registries and by better exposure assessment techniques which adequately address
both current and past exposures, including lung-retained fibre analysis where possible, as long as
performance of the latter is not made an excuse for inaction. The latter is self-evident, and can easily be
achieved through acquisition of lung samples from previously taken therapeutic lung biopsies and
resections, routine autopsies from hospitals, or from coroners’ departments. In addition, there are many
already-analyzed series of residents of given communities which may be found in the literature and used
for this purpose. Two examples of autopsy series used for exposure assessment via lung-retained fibre
analysis are our analysis of the lungs of children for US EPA [44] and our assessment of Canadian
background lung asbestos values by obtaining right upper lobe samples from coroners in nine of the ten
provinces and both Territories [45]. Examples from a wide range of patients from many states were
provided in both study subjects and controls for exposure assessment related to the man-made mineral
fibre industry [46]. Direct assessment in relation to proximity to asbestos sources has shown that lung
content varies inversely with distance from chrysotile mines and directly with duration lived within ten km
of such mines [15,20,21]. All of the latter studies were funded through peer review by public agencies in
Canada and the USA, so there is every reason to believe the same could be achieved for studies relating to
proximity to NOA.
4.2.1. Translating NOA exposure assessment to risk assessment for mesothelioma. While better cancer
registries will provide better characterization of where disease occurs, and lung-retained fibre analysis (as
well as the type of activity-based air sampling practiced by US EPA in El Dorado Hills, California [35])
will help in exposure assessment, linking the two in risk assessment is no simple matter. Cancer registries
are by definition ecological in nature, ―counting cases‖ in given areas and comparing them to national or
other rates. They cannot produce individual risk data, and the same deficiency is evident for studies such
as that of Pan et al. which ultimately rely on cancer registries for their disease data [19]. The confounding
effect of legacy exposures, particularly those related to occupation which still cause the highest proportion
of cases nation- and world-wide, are a particular problem. Only dedicated studies which provide
individual measures of both disease and exposure are likely to provide definitive answers.
4.3. ―Current‖ legacy exposures and NOA exposures: A common problem.
As past occupational exposures become an increasingly less important source of mesothelioma risk with
increasing time, current NOA exposures and legacy exposures which are due to unremediated waste or
scrap left over from the past will become the source of an increasing proportion of mesothelioma risk.
Although their origins are quite different, their effects are likely to be similar, depending on exposure
Inhaled Particles X, (23–25 September 2008, Manchester) IOP PublishingJournal of Physics: Conference Series 151 (2009) 012008 doi:10.1088/1742-6596/151/1/012008
11
patterns. The most important task we face is to identify them, as has been the case in Jefferson Parish,
Louisiana and El Dorado County (and the rest of the Western Slope of the Sierras). Once identified
however further tasks await. Once found, sources must be characterized; some will be serious and require
remediation or other corrective measures. Those which provide disease risk are of particular concern, but
we should not wait until disease has been established to begin to prevent exposures. Epidemiology is a
useful tool for identifying past ―epidemics‖, whether of infectious diseases or cancers caused by toxic
materials. But we cannot wait for the comfort of epidemiological proof where exposures we know to be
analogous to those which have produced disease in the past exist. Once identified, both legacy exposures
and NOA exposures must be eliminated as possible causes of future disease.
Acknowledgments
Funding of the lung-retained fibre analyses reported in Table 1 for the Jefferson Parish cases was provided
by the law firms who referred the cases to one of us (BWC). Funding of the lung-retained fibre analyses
for the El Dorado county pets reported in Table 2 was provided directly by the authors with assistance
with resources from their university departments, and additional assistance from Mr. Bryan Burnett of
San Diego and Mr. Terry Trent, also of California. We also thank Mr. Trent and Dr. Marci R. Culley of
Georgia State University for invaluable assistance in understanding the dynamics and time course of the
events in El Dorado County which precipitated the interest and public health concerns there from 1998 to
the present. Mr. Burnett and Mr. Trent are also co-authors of a previous abstract of preliminary results in
four of the six pets reported in Table 2 [47].
Competing Interests
Drs. Case and Abraham have acted as expert witnesses for law firms representing defendants and/ or
plaintiffs in asbestos litigation and/or compensation board proceedings in the United States and Canada,
and have been paid and unpaid consultants to regulatory agencies and compensation boards in North
America. Both have also provided unpaid assistance to community groups and residents having concerns
with the types of asbestos exposure discussed in this paper, including residents of El Dorado County,
California.
References
[1] Enterline P E and Henderson V L 1987 Geographic patterns for pleural mesothelioma deaths in the
United States, 1968-81 J Natl Cancer Inst 79 31-7
[2] Connelly R R, Spirtas R, Myers M H, Percy C L and Fraumeni J F, Jr. 1987 Demographic patterns
for mesothelioma in the United States J Natl Cancer Inst 78 1053-60
[3] Weill H, Hughes J M and Churg A M 2004 Changing trends in US mesothelioma incidence Occup
Environ Med 61 438-41
[4] Price B and Ware A 2004 Mesothelioma trends in the United States: an update based on
Surveillance, Epidemiology, and End Results Program data for 1973 through 2003 Am J
Epidemiol 159 107-12.
[5] Spirtas R, Heineman E F, Bernstein L, Beebe G W, Keehn R J, Stark A, Harlow B L and Benichou
J 1994 Malignant mesothelioma: attributable risk of asbestos exposure Occup Environ Med 51
804-11
[6] Pinheiro G A, Antao V C, Bang K M and Attfield M D 2004 Malignant mesothelioma surveillance:
a comparison of ICD 10 mortality data with SEER incidence data in nine areas of the United
States Int J Occup Environ Health 10 251-5
[7] Clifton R A. 1982 Asbestos Minerals Yearbook Centennial Edition 1981 Washington: United
Inhaled Particles X, (23–25 September 2008, Manchester) IOP PublishingJournal of Physics: Conference Series 151 (2009) 012008 doi:10.1088/1742-6596/151/1/012008
12
States Bureau of Mines pp 103-12.
[8] Rushworth D H 2005 The Navy and Asbestos Thermal Insulation Naval Engineers Journal Spring
35-48
[9] Roggli V L and Vollmer R T 2008 Twenty-five years of fiber analysis: what have we learned? Hum
Pathol 39 307-15
[10] Maule M M, Magnani C, Dalmasso P, Mirabelli D, Merletti F and Biggeri A 2007 Modeling
mesothelioma risk associated with environmental asbestos exposure Environ Health Perspect
115 1066-71
[11] Van Gosen B S 2005 Reported Historic Asbestos Mines, Historic Asbestos Prospects, and Natural
Asbestos Occurrences in the Eastern United States: U. S. Geological Survey Open-File Report
2005-1189, 1 plate.
[12] Van Gosen B S 2006 Reported Historic Asbestos Prospects and Natural Asbestos Occurrences in
the Central United States: U. S. Geological Survey Open-File Report 2006-1211, 1 plate.
[13] Van Gosen B S 2007 Reported Historic Asbestos Mines, Historic Asbestos Prospects, and Natural
Asbestos Occurrences in the Rocky Mountain States of the United States (Colorado, Idaho,
Montana, New Mexico, and Wyoming): U. S. Geological Survey Open-File Report 2007-1182,
1 plate.
[14] Van Gosen B S 2008 Reported Historic Asbestos Mines, Historic Asbestos Prospects, and Natural
Asbestos Occurrences in the Southwestern United States (Arizona, Nevada, and Utah): U. S.
Geological Survey Open-File Report 2008-1095, 1 plate.
[15] Case B W 1991 Health effects of tremolite. Now and in the future Ann N Y Acad Sci 643 491-504
[16] Sullivan P A 2007 Vermiculite, respiratory disease, and asbestos exposure in Libby, Montana:
update of a cohort mortality study Environ Health Perspect 115 579-85
[17] McDonald J C, Harris J and Armstrong B 2004 Mortality in a cohort of vermiculite miners exposed
to fibrous amphibole in Libby, Montana Occup Environ Med 61 363-6
[18] Case B W 2006 Libby Mesothelioma Update: Occupational and Non-Occupational. Lung Cancer
54 S10
[19] Pan X L, Day H W, Wang W, Beckett L A and Schenker M B 2005 Residential proximity to
naturally occurring asbestos and mesothelioma risk in california Am J Respir Crit Care Med 172
1019-25
[20] Case B W and Sebastien P 1987 Environmental and occupational exposures to chrysotile asbestos: a
comparative microanalytic study Arch Environ Health 42 185-91
[21] Case B W and Sebastien P 1989 Fibre levels in lung and correlation with air samples IARC Sci Publ
90 207-18
[22] McDonald J C, Armstrong B, Case B, Doell D, McCaughey W T, McDonald A D and Sebastien P
1989 Mesothelioma and asbestos fiber type. Evidence from lung tissue analyses Cancer 63
1544-7
[23] Case B W 1994 Biological indicators of chrysotile exposure Ann Occup Hyg 38 503-18, 410-1
[24] Case B W and Dufresne A 1997 Asbestos, asbestosis, and lung cancer: observations in Quebec
chrysotile workers Environ Health Perspect 105 Suppl 5 1113-9
[25] Nayebzadeh A, Case B W, Masse J and Dufresne A 2006 Mineralogical and exposure determinants
of pulmonary fibrosis among Quebec chrysotile miners and millers Int Arch Occup Environ
Health 79 227-36
[26] Wright R S, Abraham J L, Harber P, Burnett B R, Morris P and West P 2002 Fatal asbestosis 50
years after brief high intensity exposure in a vermiculite expansion plant Am J Respir Crit Care
Med 165 1145-9
[27] Abraham J L, Burnett B R and Hunt A 2002 Quantification of non-fibrous and fibrous particulates
Inhaled Particles X, (23–25 September 2008, Manchester) IOP PublishingJournal of Physics: Conference Series 151 (2009) 012008 doi:10.1088/1742-6596/151/1/012008
13
in human lungs: Twenty-year update on peumoconiosis database. Ann Occup Hyg 46(S1) 397-
401
[28] Abraham J L. Analysis of fibrous and non-fibrous particles. 2006 In: Rom W and Markowitz S,
eds. Environmental and Occupational Medicine. 4th ed. Philadelphia: Lippincott Williams and
Wilkins 2006 pp 277-97.
[29] Peters E S, Schmidt B A, Andrews P A, Hsieh M, Mumphrey B, Wu X and Chen V W 2007 Cancer
in Louisiana, 2000-2004 New Orleans: Louisiana Tumor Registry 91 pages
[30] Hughes J M, Weill H and Hammad Y Y 1987 Mortality of workers employed in two asbestos
cement manufacturing plants Br J Ind Med 44 161-74
[31] Martin J J 1996 Memorandum: Request for a time-critical removal action and an exemption from
the $2 million statutory limit at the Westbank asbestos site Marrero, Jefferson Parish, Louisiana
Dallas, Texas: US Environmental Protection Agency Region 6 14 pages.
[32] NIOSH 2003 Malignant Mesothelioma. Work-Related Lung Disease Surveillance Report 2002.
Cincinnati Ohio Department of Health and Human Services pp157-68.
[33] NIOSH 2008. Malignant mesothelioma: Counties with highest age-adjusted death rates (per million
population), U.S. residents age 15 and over, 2000–2004. Ref. No. 2007T07-10, Work-Related
Lung Disease (WoRLD) Surveillance Report (update), Surveillance Branch, Division of
Respiratory Disease Studies, NIOSH, Morgantown, WV 26505. June 23, 2008. Available online
at http://www2.cdc.gov/drds/WorldReportData/pdf/2007T07-10.pdf (Accessed October 5,
2008).
[34] Davis J M, Addison J, McIntosh C, Miller B G and Niven K 1991 Variations in the carcinogenicity
of tremolite dust samples of differing morphology Ann N Y Acad Sci 643 473-90
[35] US Environmental Protection Agency Region 9 2005 U.S. EPA Asbestos Assessment for El Dorado
Hills. (San Francisco, CA: US Environmental Protection Agency Region 9). 6 pages.
[36] Meeker G P, Lowers H A, Swayze G A, Van Gosen B S, Sutley S J and Brownfield I K. 2006
Mineralogy and Morphology of Amphiboles Observed in Soils and Rocks in El Dorado Hills,
California. (Reston, VA: U.S. Geological Survey) Open-File Report 2006-1362 47 pages
[37] Buckingham D.A. and Virta R.L. U.S. Geological Survey 2007 Asbestos Statistics in: Kelly, T.D.,
and Matos, G.R., comps., Historical statistics for mineral and material commodities in the
United States: U.S. Geological Survey Data Series 140. Available online at
http://minerals.usgs.gov/ds/2005/140/asbestos.pdf (Accessed June 26, 2008)
[38] McDonald J C, Harris J M and Berry G 2006 Sixty years on: the price of assembling military gas
masks in 1940 Occup Environ Med 63 852-5.
[39] Hodgson J T and Darnton A 2000 The quantitative risks of mesothelioma and lung cancer in
relation to asbestos exposure Ann Occup Hyg 44 565-601
[40] Peto J, Seidman H and Selikoff I J 1982 Mesothelioma mortality in asbestos workers: implications
for models of carcinogenesis and risk assessment Br J Cancer 45 124-35
[41] Price B and Ware A 2005 Mesothelioma: risk apportionment among asbestos exposure sources Risk
Anal 25 937-43
[42] Goldberg M, Imbernon E, Rolland P, Gilg Soit Ilg A, Saves M, de Quillacq A, Frenay C,
Chamming's S, Arveux P, Boutin C, et al. 2006 The French National Mesothelioma Surveillance
Program Occup Environ Med 63 390-5
[43] Yeung P and Rogers A 2001 An occupation-industry matrix analysis of mesothelioma cases in
Australia 1980-1985 Appl Occup Environ Hyg 16 40-4
[44] Case BW K M, Harrigan M, Dufresne A. 1994 Lung fibre content of American children aged 8-15
years. Ann Occup Hyg 38 639-45
[45] Case B W, Sebastien P and McDonald J C 1988 Lung fiber analysis in accident victims: a
Inhaled Particles X, (23–25 September 2008, Manchester) IOP PublishingJournal of Physics: Conference Series 151 (2009) 012008 doi:10.1088/1742-6596/151/1/012008
14
biological assessment of general environmental exposures Arch Environ Health 43 178-9.
[46] McDonald J C, Case B W, Enterline P E, Henderson V, McDonald A D, Plourde M and Sebastien P
1990 Lung dust analysis in the assessment of past exposure of man-made mineral fibre workers
Ann Occup Hyg 34 427-41
[47] Abraham J, Case BW, Burnett B, Trent T. 2005 Lung-Retained Fiber in Pets Confirms
Environmental Exposure to Naturally-Occurring Asbestos in Western El Dorado County,
California. Am J Resp Crit Care Med 171(S1) A818.
Inhaled Particles X, (23–25 September 2008, Manchester) IOP PublishingJournal of Physics: Conference Series 151 (2009) 012008 doi:10.1088/1742-6596/151/1/012008
15